Sources for the study of deserted medieval settlements – medieval taxation records

Nothing is certain but death and taxes…. or so the saying goes… and though we all may dislike paying tax, the record of such payments afford interesting insights into many aspects of everyday life. The wealth of taxes raised by the crown throughout the medieval period can provide an indication of population size and wealth as well as some of the names of the population.

Saying that though, as with all the sources of information about medieval population, it is never that easy. Taxation records do not necessarily record the total population, just those with enough assets to be taxed. Every taxation was set out with its own criteria and these might not apply to everyone, and as in the present day there was much avoidance – this is nothing new. Many of the tax records for the medieval period can be found in the National Archives in Kew. Many of them are located in the section E179 – also known as the ‘King’s Remembrancer, particulars of account and other records relating to lay and clerical taxation’. An online search facility allows you to search for a place-name and to show which taxations records are present for. This does not give you the amount but its presence on the document.

Another factor that needs considering is the survival of records, not all taxation records survive for all counties and not all areas are taxed in the first place. This means getting a full picture of the entire country can be difficult. The records chosen by the Beresford’s Lost Villages website are those that are available in published form and provide a complete a record as possible for the entire country. They stretch from the early fourteenth century through to the sixteenth century thus covering the key period when most settlements should have been in existence to be classed as a medieval village but also into the period where a decline may have started. It is unfortunate that no real nation-wide survey is available between Domesday (1086) and the early fourteenth century although on occasions they are available for small regions. Finally when a place-name is mentioned in the taxation what does it represent? In some regions this will be a single nucleated village, others it may be the name of a village which also includes other smaller hamlets and outlying settlements. Then in other areas it may a region with no nucleated settlement at all. A place-name in the taxation records does not necessarily equate to a nucleated village on the ground.

The main taxations used are the Lay Subsidy taken in 1344, Poll Taxes in 1377, 1379 and 1381, Lay Subsidies in 1524, 1525 and 1543, a record of number of households in each diocese in 1563.

Lay Subsidy of 1334
The Lay Subsidy of 1334 records one of the special taxes granted by Parliament to the Crown to help with the extra expenses incurred arising from continued troubles with France and Scotland (Glasscock 1975). The tax was upon personal wealth, the value of an individual’s movable goods rather than their land and buildings, and was applied only to the laity (Glasscock 1975). Different payment fractions occurred in different subsidies, but that for the 1334 subsidy was levied at a fifteenth of the value from rural areas and as a tenth of the value from the boroughs and areas of ancient demesnes (Glasscock 1975). This followed not long after a 1332 tax which used the same fractions, but the 1334 tax differed in the fact that the sum paid was a figure that was agreed on by the local community involved. The amount could not be less than that paid in 1332 and was negotiated between the community and the appointed tax officials (Glasscock 1975).

Many subsequent taxes were based on the 1334 figures, and so lose their value in assessing any changes that have occurred in the population; as Glasscock (1975: xvi) notes: ‘within a very short time the tax ceased to bear any direct relationship to the lay wealth and taxable capacity of the country’. However, due to the fact that any gaps in the 1334 record can be filled by any subsequent records that were based on the 1334 tax, this lay subsidy is one of the most complete records of taxation from the fourteenth century for the country as a whole. There were a number of items that were not taxable in the 1334 subsidy, including clerical property that was included in the 1291 taxation of Pope Nicholas, as well as locally argued exemptions, mainly relevant to the Cinque Ports (Glasscock 1975). Although this record does not provide population numbers, or exact wealth, it provides an overview of possible surplus wealth of the region in the fourteenth century and can be used to compare the fortunes of different settlements.

Poll Taxes of 1377, 1379 and 1381
During the later fourteenth century, fifteenths and tenths (subsidies) were still collected as in 1334, but experiments took place at attempting to tax individuals rather than their wealth. The Poll Tax of 1377 introduced a per capita levy, which was to be repeated in 1379 and 1381, although by this time the old habits were returning, with the collectors instructed to base the amount collected on the wealth of the individuals (Fenwick 1998). The 1377 Poll Tax was collected from every layman and woman over the age of fourteen who was not a mendicant, at a rate of one groat (fourpence) (Fenwick 1998). The Poll Tax of 1379 introduced a scale of payments from fourpence to ten marks, to be collected from every lay married and single man and every single woman over the age of sixteen (Fenwick 1998). The Poll Tax of 1381 was collected from every lay man and woman of fifteen years and over at a rate of three groats (one shilling), but in order that the rich should help the poor, everyone was ‘charged according to his means’, with the resulting total number of shillings from any one vill being equivalent to the number of taxpayers (Fenwick 1998: xvi). The clergy were exempt from the Parliamentary taxes, although there was an attempt to remove the other exemptions that had previously existed for the lay subsidies, but with varying degrees of success (Fenwick 1998).

The Poll Tax of 1377 did not require detailed records to be kept of individuals who were taxed; only the numbers of individuals and the amount due were noted, but more detailed records were kept for the taxes of 1379 and 1381 (Fenwick 1998). The survival of the poll tax documents varies across the country.  Two values are recorded – the number of individuals taxed and the total amount. All the surviving records have been published in three volumes by Caroline Fenwick (Fenwick 1998, 2001, 2005).

Lay Subsidies of 1524, 1525, and 1543
The collection of the rates as set down by the Lay Subsidy of 1334 continued until 1623 and became known as ‘the fifteenth’ (Hoyle 1994). During the Tudor period the lay subsidy was revived, running alongside that of the fifteenth. This new subsidy was based on either income from land, other income, or goods or wages, with each payee only being taxed on one category (Hoyle 1994). Early records of the Tudor lay subsidies contain little information on the amounts taxed at certain villages – they only show the amount collected by different tax collectors. In 1523 an act was passed allowing four lay subsidy payments in the subsequent years. The records were expanded to include information on which towns, parishes or other taxation unit had provided which amount. As such, in the first payment of the four in 1524, the information on individual taxpayers was included (Hoyle 1994, Sheail 1998a). The subsidies of 1524 and 1525 recorded every man who was worth more than £1. The later two surveys changed the criteria to men who earned more than £50 in a year (in 1526) and those who were rich in moveable goods (in 1527). The rates at which individuals were taxed depended on a number of factors, shown below.

1s in the pound: Annual income of land and other sources
1s in the pound: Capital value of moveables worth £20 and upward
6d in the pound: Capital value of moveables worth £2 and upward to £20
4d in the pound: Capital values worth £1 and under £2
4d paid: Those aged 16 years and above and who earned wages of and in excess of £1 a year

The records for the 1524/25 subsidies are one of the most complete and useful surveys, and Sheail (1998a, 1998b) has attempted to produce calculations from the different surveys of the period to provide combined figures for individual vills. As with all the tax records, there are numerous problems using the data, such as the omission of much of the clergy and those valued at under £1, but as Sheail (1998a: 36) concludes, the study of the survey of specific counties and hundreds overcomes some of the issues as ‘they were surveyed by the same men and consequently have a greater degree of uniformity’.

For a number of counties Sheail also presents the figures from subsidies in the 1540s (from 1543, 1544 or 1545) but only the number of people assessed, not the amount of money requested. Although this later survey has often been cited as being superior to the earlier 1524/25 surveys, the record is not complete and is much damaged (Sheail 1998a).

Diocesan Returns of 1563
The Privy Council commanded in 1563 that every bishop should provide an outline of the administrative structure of their dioceses including a list of the number of households in every village and hamlet in each parish (Dyer and Pallister 2005). It is not fully understood why these data were required but suggestions include the idea of quantifying the number of potential almsgivers available for contribution to poor relief (Dyer and Pallister 2005).

Although all 26 dioceses returned an initial report of the administrative structures, they did not have to reply instantly with the number of households. So unfortunately some returns have only survived in partial form and have no lists of households available. Only 12 dioceses have the list of households surviving, and it has been suggested that the missing ones were lost en masse at some point in the past (Dyer and Pallister 2005). Records survive for Bath and Wells, Canterbury, Carlisle, Chester, Coventry and Litchfield, Durham, Ely, Gloucester, Lincoln and Worcester as well as the Welsh dioceses of Bangor and St David’s. There are no records of households from Bristol, Chichester, Exeter, Hereford, Llandaff, London, Norwich, Oxford, Peterborough, Rochester, St Asaph, Salisbury, Winchester and York. On some occasions a later record for 1603 does survive. As with all such records there are instances of rounding of numbers, errors in copying and other general issues. Tests of comparison with other localised records of households from around the same period show a surprising similarity which confirms the usefulness of these records (Dyer and Pallister 2005).

These are just a few of the sources that can be used to study medieval settlements and future posts will look at other sources.


Dyer, A. and D.M. Palliser 2005. The Diocesan Population Returns for 1563 and 1603. Oxford: Oxford University Press.

Fenwick, C.C. 1998. The Poll Taxes of 1377, 1379 and 1381, Part 1, Bedfordshire-Lincolnshire. Oxford: Oxford University Press.

Fenwick, C.C. 2001. The Poll Taxes of 1377, 1379 and 1381, Part 2, Lincolnshire-Westmorland. Oxford: Oxford University Press.

Fenwick, C.C. 2005. The Poll Taxes of 1377, 1379 and 1381, Part 3, Wiltshire-Yorkshire. Oxford: Oxford University Press.

Glasscock, R.E. 1975. The Lay Subsidy of 1334. London: British Academy.

Hoyle, R. 1994. Tudor Taxation Records: a guide for users. London: PRO Readers’ Guide 5.

Sheail, J. 1998a. The Regional Distribution of Wealth in England as Indicated in the 1524/5 Lay Subsidy. Volume 1. London: List and Index Society Special Series 28.

Sheail, J. 1998b. The Regional Distribution of Wealth in England as Indicated in the 1524/5 Lay Subsidy. Volume 2. London: List and Index Society Special Series 29.

Sources for the study of deserted medieval settlements – Domesday Book

One of the first times many settlements are recorded is in 1086 and the Domesday Book. But it is not that simple….. The Domesday Book can be difficult to interpret and does not directly record villages but manors – areas of landholdings that may, or may not contain a nucleated settlement, or several dispersed (and not separately recorded) settlements. A number of manorial names also cannot be traced down to the modern day – are these deserted settlements that where depopulated so early (not long after Domesday) that they fail to be recorded in the later records, or has such a dramatic change in name occurred with no account of the change? This blog looks at the evidence from the Domesday Book that has been used by the Beresford’s Lost Villages website, examines some of the challenges this presents and gives some examples of the complexity of this record.

There are several different transcriptions of the Domesday Book. The one that has been used by the website is that published by Philimore. This is available as printed-copy but it has also been digitised by the Domesday Book Project (see The computer files and data from the project are available for download from the University of Hull data repository.

The Domesday Book was compiled on the order of William I in 1086. The data in the Domesday Book are recorded by landowner, then manor. As the basic unit of measurement, the manor was never defined in the Domesday record. It does not directly equate to a settlement, but more to a unit of land, however it is usually linked to the vill (village) in which it was located. The result is that each vill may constitute part of several different manors, and therefore have more than one record in Domesday. These separate manors within the same vill may also have a number of different landowners. Land in outlying areas may also be included under the name of the manor to which it is attached. This can mean that land and people may be recorded as located in one particular area but might actually be located at a distance. Land under one entry within Domesday may also record several vill names.

Here is a typical example of a Domesday entry for the deserted settlement of Lasborough in Gloucestershire.

‘Hugh also holds Lasborough from the Bishop himself. Leofwin held it. 5 hides, In lordship 1 plough; 5 villagers and a priest with 2 ploughs. 7 slaves. The value was £10; now 50s.’ (Moore 1982: 30,2)

Domesday Book Gloucestershire entry for Lasborough
Domesday Book Gloucestershire entry for Lasborough

The main unit of measurement used in the Domesday Book as a whole was the hide. The Saxon hide was a theoretical unit of land required to support a family farmstead, but by the Late Saxon period, it had become a unit of taxation and hence was a fiscal unit rather than signifying a specific area of land. So at Lasborough above the taxation value was 5 hides. In the old area of Danelaw (much of northern England ) the measurements are given in carucates, the Danelaw equivalent to the hide.

A number of different aspects of the landholding were recorded in the survey. These include the number of ploughlands and a formula referred to as ‘land for x ploughs’. The number of ploughs that are recorded is thought to signify the amount of agriculture being practiced at the time of the survey. The ‘land for x ploughs’ provides us with a figure of the agricultural potential of the manor, although the actual amount of land that was farmed may be different as is signified by the differences in the ‘land for x ploughs’ and the number of ploughs recorded in each manor. The meaning of the ‘land for x ploughs’ value has been much debated (Harvey 1985, 1987, Higham 1990).

The to-be deserted settlement at Brackenborough in Lincolnshire shows the complexity of the Domesday record. There are two entries for Brackenborough, both in the lands of Alfred of Lincoln. The first entry reads:

‘In Brackenborough 1 bovate of land taxable. land for 2 oxen. Ranulf, Alfred’s man has 1 plough. 4 villagers with 1/2 plough. A jurisdiction of Alvingham. Meadow, 10 acres’ (Morgan and Thorn 1986: 27,23)

The second entry follows straight after:

‘In this village Eadric and Hoc had 6 bovates of land taxable. Land for 14 oxen. Ranulf, Alfred’s man, has 1 villager and 10 freemen with 2 ploughs. 4 parts of a mill, 2s; meadow, 18 acres. Value before 1066, 16s; now 40s’ (Morgan and Thorn 1986: 27,24)

As Brackenborough is in Lincolnshire, here we have recording in the Danelaw equivalents – and in this case bovates – a sub-division of carucates – there were 8 bovates to a carucate. The first entry is an area of land at Brackenborough but belonging to the manor of Alvingham. The second entry is for the manor of Brackenborough. So altogether at Brackenborough there are 7 bovates of taxable land recorded. The ‘land for x ploughs’ in both these cases is also a smaller division – one plough was equivalent to 8 oxen so in the first entry there in essence is land for 1/4 plough and in the second entry land for 1 3/4 ploughs – so in total land for two ploughs. However it is clear that in the first entry there are more ploughs than land for ploughs as 1 1/2 are recorded. In total the land had the potential (land for x ploughs) of 2 ploughs but there were actually 3 1/2 ploughs in action.

Brackenborough Deserted Medieval Village, Lincolnshire. Copyright Google Earth.
Brackenborough Deserted Medieval Village, Lincolnshire. Copyright Google Earth.

As can be seen from the examples above other resources recorded at Domesday include population. The population of each manor is recorded as numbers of different classes of population which do vary in terminology and meaning in different regions. In the Phillimore translations villagers are equivalent to villeins, freemen are equivalent to sokemen and smallholders to bordars in other translations. Other forms of population recorded include burgesses, cottagers, slaves and priests. A villager was a member of the vill with certain burdens and responsibilities. A freeman was free from many of the burdens that rested on a villager. A smallholder had less status and land than a villager.  Also recorded in Domesday are resources such as mills, meadow, wood, woodland pasture, underwood, marsh, saltpans, livestock, fisheries etc. Again they may not be whole items. So at Brackenborough above there was ‘4 parts of a mill’ – how much this equates to is not known, if the mill was at Brackenborough or is a share of a mill elsewhere is not clear.

There has been much debate over the nature of the record presented by Domesday. For instance, the record can hide or omit settlement and population; it has been noted that the number of actual tenants in 1086 may in fact be 50% more than those recorded if a similar number of sub and joint tenancies were present in the late eleventh century as are recorded for the thirteenth (Postan 1972). Nevertheless, Domesday provides a region-wide record taken at a specific point in time which can be used to assess the extent, if not the true nature, of settlement in the eleventh century.

Lost Villages Database contents

Two items are recorded on the Beresford’s Lost Villages website: a reference to the entries in the Domesday Book Phillimore editions, and the minimum number of individuals that are recorded as belonging to that manor. The data for these sections were derived from the Domesday Explorer Project which was based at the University of Hull.

The Phillimore county editions number each of the entries in the record with a coding system. Each county edition is divided into the respective landowners, with the first (usually the King) given the number 1, and the second landowner, 2 etc. For each entry under a particular landowner, the entries are given separate numbers. So entry 2,5 would be landowner 2 and entry 5. This form of notation has been used by the website so people can be directed straight to the entry or entries about a particular manor. These details can be found in the printed versions or via the web link to the databases at the University of Hull. The Phillimore entry is prefixed with a county code to identify which county record should be used. Some settlements have since changed counties so do not appear in the pre-1974 county in which they are now placed. On a number of occasions land outside one county will be recorded in a different county.

A total minimum population figure for a manor was calculated based on the number of villagers, freemen, smallholders, cottagers, slaves, burgesses and priests. The total for each settlement has been calculated from the individual manors with the same place-name recorded under different landowners in the Domesday Book – hopefully giving an indication of a minimum population for a manor – and hopefully an indication of a size of a settlement – if this was a nucleated village at the centre of the manor – but remember that the record may not be directly recording a single nucleated settlement.

For each site that has a Domesday record a direct link is provided on the website to the Open Domesday website that provides a summary of the information, location map and a picture of the original Domesday entry.

Further Information

There are a number of online resources that may be of help to people looking at the Domesday Book:

Domesday Explorer Project –

Open Domesday –

Hull University Repository –



Harvey, S.P.J. 1985. Taxation and the ploughland in Domesday Book, in P. Sawyer (ed.) Domesday Book. a reassessment: 86-103. London: Edward Arnold.

Harvey, S.P.J. 1987. Taxation and the economy, in J.C. Holt (ed.) Domesday Studies: 249-64. Woodbridge: Boydell Press.

Higham, N.J. 1990. Settlement, land use and Domesday ploughlands. Landscape History 12: 33-44.

Moore, J.S. 1982. Domesday Book Gloucestershire. Chichester: Phillimore.

Morgan, P. & C. Thorn 1986. Domesday Book Lincolnshire. Chichester: Phillimore.

Postan, M.M. 1972. The Medieval Economy and Society. Harmondsworth: Penguin.


Problems identifying villages – the cases of Roel and Manless Town

This week’s blog looks at issues with identifying deserted villages with two examples of complex cases – Roel and Manless Town – both in Gloucestershire.

Lists of deserted medieval villages are compiled from a variety of sources and evidence. Sometimes there is a range of documentary sources recording a settlement with a specific name and listing villagers, tax payments, land under plough, duties owed etc. On some occasions there may be extensive earthworks signifying former houses and route ways. Matching both these sets of evidence can on some occasions seem fairly simple – the existence of a church, the presence of the place-name still in existence for a farm, or the settlement is clearly recorded on a map. However even the most straight forward of cases can become more and more complex once the intricacies of the evidence are explored. One example that demonstrates this is Roel located in the Cotswolds in Gloucestershire. A different dilemma is faced by the site of Manless Town also in Gloucestershire where local legend and more than 5 different place-names make the settlement difficult to identify in the documentary records.


The village of Roel has been the subject of in-depth study combining the archaeological and documentary evidence to good effect (Aldred and Dyer 1991). It was this research that uncovered the complex development of Roel and its neighbouring village, Hawling. Part of the settlement of Roel is still visible with a row of at least six crofts aligned to the east of a north-south hollow way with another fainter three crofts to the south. To the west of the hollow way the modern farm complex of Roel Farm sits over the site of the manor house, church and vicarage. However this is not the entirety of settlement evidence for the village of Roel as further crofts, located 2km to the south, to the north of Hawling village have now been identified as part of Roel settlement, named Roelside (Aldred and Dyer 1991).

Earthworks at Roel Farm. Copyright Google Earth
Earthworks at Roel Farm. Copyright Google Earth

It is thought that Roel developed as a secondary settlement of Hawling, originating as a woodland hamlet (Aldred and Dyer 1991). The church is recorded as a subordinate of Hawling in 1174. Domesday records a minimum population of 21. In 1294 28 tenants are recorded as owing services with 31 tenanted yardlands. In 1327 13 people are assessed (Franklin 1993: 68). In the fourteenth-century taxations it is included with Hawling. A below average amount is paid in 1334. The documentary evidence suggests there may have been as many as 30 households in the early fourteenth century classed as the settlement of Roel.


The physical evidence at Roel Farm, the traditional site of Roel, does not suggest many more than nine house plots  – and no where near the 30 households suggested above. Hence the further evidence of settlement to the north of Hawling was investigated. These earthworks had been considered to be evidence of settlement shrinkage at Hawling as they form a continuous extension to the village, just separated by a stream. The documentary evidence, however, has confirmed this area once formed part of Roel village, named Roelside (Aldred and Dyer 1991). Further analysis of the documentary evidence shows that the split between the two areas of settlement, always recorded singularly as ‘Roel’ saw around a quarter of the population located at Roel and three quarters at Roelside, 2 km away (Aldred and Dyer 1991).

Earthworks at Roelside, Hawling, Gloucestershire. Copyright Google Earth
Earthworks at Roelside, Hawling, Gloucestershire. Copyright Google Earth

Roel and Hawling

So the documentary evidence tells the tale of a single settlement unit of Roel. And tradition therefore equated that with the settlement evidence at Roel Farm. But in fact the settlement of Roel was actually formed of two separate groups of dwellings 2 km apart – Roel and Roelside. Roel was always the smaller of the two, but seen as the centre of the settlement, with the manorial centre and church. Roelside sat cheek by jowl with the village of Hawling, so much so that only a stream separated a villager of  Roelside with a villager in Hawling. The decline of the village of Roel effects not only the settlement at Roel Farm but also the settlement at Roelside, where as Hawling survives.

Roel declines in the later middle ages, with 15 tenants recorded in 1355 (Aldred and Dyer 1991). As the number of tenants decreased, land was engrossed into larger units, but these were not successful and the tenants left. By the 1460s the last tenants relinquished land (Aldred and Dyer 1991).  Roel and Roelside seemed to decline at the same time. The decrease in the number of tenants appears as a result of outward migration, possibly due to poor yields, rather than forced eviction. It is postulated that the reason that Hawling survived but Roel became deserted can be seen in the diversity of social structure present at Hawling and better agricultural production.

Manless Town

The settlement in the fields known as ‘Manless Town’ to the south of Birdlip also pose an interesting question – what is the name of this settlement? The remains of a medieval settlement and a Roman site can be seen on either side of the Climperwell to Caudle Green road but linking these remains to a documented settlement is not easy (Smith 1998). The archaeological evidence is not clear cut – and the Roman nature of some of the remains is debated. In 1962 trial trenches were placed across the site to test for remains below the surface (Wingham and Spry 1993). These revealed a range of features including stone walls and rubble surfaces however there has been disagreement over their interpretation with the original excavator favouring an interpretation in the Roman period, and later authors suggesting a medieval date (Wingham and Spry 1993).  In 1992 the remaining earthworks were plotted and field walking produced medieval pottery dating to the twelfth to thirteenth centuries as well a range of Roman material (Smith 1998).

Manless Town settlement. Copyright Google
Manless Town settlement. Copyright Google

The physical evidence suggests medieval settlement and the name ‘Manless Town’ lends credence to the idea of a deserted settlement. However the first known reference to this field as Manless Town was in 1622 (Jurica 1981). The settlement has also been recorded as Haywick, Munley Towne, Old Mondley, Longlorn Town and Keywich as will be shown below (Newbury 1993). It is therefore unclear as to the original name of the settlement, and hence difficult to trace any taxation records or indications of the size or wealth of any population.

The name Manless would suggest a deserted settlement and local legends would seem to confirm this, but it has also been suggested that the place-name originates from a manorial name (Smith 1964). The documentary evidence proposes a range of names and stories. In a hand written document dating to 1677 there is the record of ‘a place called Keywich, where there was a market… ye men of which place being destroyed was called Munly Towne’ (Newbury 1993: 35). In 1731 a survey map includes the text ‘A patch of Plumb Hey within the ruins of Old Mondley formerly a Market Town and a Roman Station was Sacked and Burnt in the Wars of King John…’ (Newbury 1993: 33). On a ‘Survey of Lands in Brimspfield’ dating to the late eighteenth century a note records ‘stood Longlorn Town, which was destroyed in the reign of King John, then and still traces of Foundations to be seen and it has since that Time been called Manless Town’ (Newbury 1993: 33). In 1779 it is recorded as a hamlet by Samuel Rudder but with the caveat ‘if a place can be called so with no house in it’ (Rudder 1779: 310). He also mentions that the original name for this area was Haywick, the location where a weekly market was held in the reign of Edward III, but the men of the settlement were all killed and since then it has been known as Manless Town. So here we have the blame placed at the feet of King John, possibly Edward III and ideas of the settlement being destroyed and the men all killed. We have different names for the settlement given by each map or author. Unfortunately no further evidence for the settlement has been forthcoming and it remains one of those sites which has little documentary evidence although the earthworks and archaeological finds do suggest medieval occupation. The presence of a later sheepcote over the earthworks, may suggest the fate of the settlement was  a little more mundane that the antiquarian reports of mass killings, and the fate may have been similar to other small settlements in the area which were gradually depopulated of people and repopulated with sheep.


So what do the examples of Roel and Manless Town tell us about the study of deserted settlement? With Roel it indicates that the documentary evidence for a settlement may represent physical evidence across a wide area and that the earthworks next to a continuing settlement may not necessarily represent a contiguous part of that settlement. With Manless Town we have been shown that clear physical evidence and tales of deserted settlement may not lead to an easily identifiable settlement in the documentary record. Identifying deserted settlements on the ground is one thing – linking these remains to the documentary record is another challenge.


Aldred, C. and C. Dyer 1991. ‘A Medieval Cotswold Village: Roel, Gloucestershire’, Transactions of the Bristol and Gloucestershire Archaeological Society 109: 139-170.

Franklin, P. 1993. The Taxpayers of Medieval Gloucestershire. Stroud: Alan Sutton

Jurica, A.R.J. 1981. ‘Brimpsfield’, in N.M. Herbert (ed.) A History of the County of Gloucester. Volume 7: 140-150. Oxford: Oxford University Press.

Newbury, J.R. 1993. ‘Map and Documentary Interpretations in Brimpsfield Parish’, Glevensis 27: 33-35.

Rudder, S. 1779. A New History of Gloucestershire. Cirencester: S Rudder.

Smith, A.H. 1964. The Place-Names of Gloucestershire Part 1. Cambridge: Cambridge University Press.

Smith, N. 1998. ‘Manless Town, Brimpsfield: An Archaeological Survey’, Glevensis 31: 53-58.

Wingham, H. and N. Spry 1993. ‘More Recent Views on Manless Town, Brimpsfield SO 928 116’, Glevensis 27: 26-32.